Background
Colorectal cancer is the third most common cancer worldwide [
1], and metastasis is the major cause of cancer mortality. Small invasive cells at the advancing edge of tumors are recognized as an independent prognostic factor for colorectal cancer [
2]. Increasing evidence has established a link between overexpression of matrix metalloproteinases (MMPs), which play critical roles in cancer cell invasion and metastasis by degrading the extracellular matrix, and cancer stage and/or prognosis [
3]. In the case of colorectal carcinomas, MMP-7 overexpression is correlated with increased invasion and metastasis. Therefore, MMP-7 overexpression is recognized as an independent marker of colorectal cancer progression [
4‐
10].
Regulation of human MMP-7 gene expression is dependent on activator protein 1 (AP-1), which is a transcription factor that binds to the promoter region of MMP-7 [
11‐
13]. In addition, MMP-7 expression via JNK/AP-1 pathway activation has been recently observed in cultured cells treated with exogenous H
2O
2 [
14,
15]. Compared to normal cells, cancer cells produce large amounts of reactive oxygen species (ROS) [
16], which are involved in multiple signaling cascades related to carcinogenesis, including invasion and migration of cancer cells [
17‐
20]. During the transition to a malignant phenotype, ROS are known to activate mitogen-activated protein kinases (MAPKs), including the ERK, JNK, and p38 pathways, leading to expression of MMPs for tumor invasion and metastasis [
21‐
23]. ROS are produced not only as byproducts of energy metabolism in mitochondria and the cytosol but also as signaling messengers upon activation of various cell membrane receptors for growth factors [
24,
25], cytokines [
26], and integrin [
27] via membrane NADPH oxidase (NOX). Alteration of MMP expression by NOX1-generated ROS has been studied in various types of cells, including K-Ras-transformed normal rat kidney cells [
28], human pancreatic cells [
29], and prostate cancer cells [
30]. However, there has been no such study performed using colon cancer cells.
The microenvironment of tumors is often characterized by an insufficient amount of nutrients available for rapid cell proliferation, leading to alteration of the AMP/ATP ratio and activation of AMP-activated protein kinase (AMPK), an energy-sensing kinase that regulates cell metabolism. Similarly, energetic stress during cancer metastasis is known to activate AMPK [
31‐
33]. Although AMPK has also been shown to be sensitive to oxidative stress [
34], its involvement in cancer metastasis is somewhat dependent on the type of stimulus and oxidative conditions. Lysophosphatidic acid-induced activation of AMPK enhances ovarian cancer cell migration [
35], whereas exogenous chemical-induced activation of AMPK inhibits melanoma and colon cancer cell migration [
36‐
38]. Increased AMPK activity has been observed in colon cancer cells in response to exogenous H
2O
2 [
39]. However, it is unclear whether or not ROS-associated AMPK activation is related to NOX1, which is highly expressed in colon cancer cells compared to normal adjacent tissues [
40]. On the other hand, AMPK has been reported to suppress phorbol ester-induced ROS production [
41], and AMPK deletion is associated with increased NOX activity [
42,
43]. Therefore, the precise regulatory relationship between AMPK and NOX needs to be clearly determined in order to develop a proper strategy to block cancer invasion and metastasis.
In the current study, we investigated the regulatory roles of NOX in MMP-7 expression and AMPK activity by comparing invasive behaviors of HT29 and SW620 colon cancer cells. To induce invasion, cells were treated with 12-
O-tetradecanoylphorbol-1,3-acetate (TPA), which has been demonstrated to induce migration and invasion in several human cancer cells via induction of MMPs [
44‐
46].
Discussion
MMP-7 is unique in that it is predominantly expressed in epithelial cells as opposed to other MMPs, which are mostly expressed in the stroma [
50,
51]. Overexpression of MMP-7 in human squamous cell carcinomas, in particular colon cancer cells, shows a strong positive correlation with metastatic potential of cancer cells [
52‐
54]. In the current study, we observed that MMP-7 was differentially expressed in colon cancer cell lines, HT29, Caco2, SW620, and HCT 116, having different invasive potentials. The TPA-induced transition from less (HT29 and Caco2 cells) to highly metastatic (SW620 and HCT116 cells) was associated with enhanced MMP-7 expression, which was dependent on the molecular switch from NOX1 to NOX2.
MMP gene expression is not constitutive but is rather regulated by various cytokines via activation of intracellular signaling pathways [
55]. In our current study, complete blockage of TPA-induced invasion of both HT29 and SW620 cells was observed in the case of either MMP-7 silencing or pretreatment with DPI and Apo (NADPH oxidase inhibitors), indicating a regulatory role for NADPH oxidase in MMP-7 induction as well as cancer cell invasion. It has been reported that MMP-7 expression increases in response to exogenous H
2O
2 [
15]. In this study, we clearly observed that the source of ROS governing MMP-7 expression in colon cancer cells was NADPH oxidase. In addition, the current results of NADPH oxidase activity measurement showing that NOX1 siRNA had no effect on TPA-induced ROS generation demonstrate that TPA-induced ROS was independent of NOX1 despite high NOX1 levels in HT29 cells. A previous study by Sadok
et al., [
56] showed that NOX1 silencing using siRNA decreases superoxide production in HT29-D4 cells. This contradictory result could be attributed to a difference in the type of stimulus given to cells, because in the previous study, cells were induced to be in migratory phase by being placed on collagen-I which activates integrin signaling. However, further study is required for the underlying mechanism to be solved. Our current results also demonstrate that treatment of HT29 cells with TPA reduced NOX1 expression to a level similar to that in untreated SW620 cells. These results are somewhat in line with previous reports that NOX1 is constitutively expressed in the colon epithelium to a similar degree as in differentiated colon cancer cells [
57], and increased NOX1 expression suppresses proliferation of cancer cells, driving them into a differentiated cancer state [
58]. An important role of NOX1 in tumorigenesis has been suggested based on several findings. Induction of NOX1 mRNA transcription has been observed in non-cancerous cells such as smooth-muscle cells, and fibroblasts by stimulation with platelet-derived growth factor or epidermal growth factor [
59,
60]. It is also reported that K-Ras mutation correlates with increased NOX1 mRNA expression and colon tumor phenotype [
61]. However, based on human tumor tissue array findings of diminished NOX1 expression at a more advanced tumor stage [
58], it is suggested that tumor-promoting action of NOX1 seems to be an early event [
62]. Our current results showing that NOX1 expression is diminished as colon cancer cells undergo invasive phenotype change also support such suggestion.
In contrast to NOX1, NOX2 expression was elevated and accompanied by enhanced MMP-7 expression in TPA-treated HT29 cells, which was similar to that in basal SW620 cells. In addition, NOX2 siRNA inhibited basal NADPH oxidase activity, TPA-induced ROS production, MMP-7 expression, and cell invasion, indicating that NOX2 plays a critical role in induction of MMP-7 expression in invasive cancer cells. These results further indicate that switch from NOX1 to NOX2 induces colon cancer cells to be highly invasive. In support of this, ROS have been reported to play signaling roles in normal cells such as cardiomyocytes and keratinocytes for pro-survival [
63] or MMP-9 induction [
64]. Likewise, the results of the current study demonstrate that MMP-7 expression was accompanied by activation of MAPKs as well as the transcription factors NF-κB and AP-1. At the same time, NOX2-activated ROS constituted an autoregulatory loop for expression of NOX2 and its regulators, p47phox and p67phox in TPA-treated colon cancer cells. This is similar to the previous finding that expression of phagocytic NADPH oxidase subunits is up-regulated by stimulus-induced ROS in a positive feedback mechanism [
65].
AMPK activation in colon cancer cells has been reported to suppress the expression of genes associated with invasion and metastasis, including integrin β1 and cyclooxygenase-2 [
37,
38]. In the current study, we observed for the first time that AMPK activity was inversely linked to MMP-7 expression in colon cancer cells. At a basal level, low MMP-7 expression and high AMPK phosphorylation were observed in less invasive HT29 cells, whereas low AMPK phosphorylation and high MMP-7 expression were observed in SW620 cells. Such an inverse relationship between AMPK activity and MMP-7 expression was further confirmed based on our observations that TPA-induced MMP-7 expression was accompanied by reduced AMPK phosphorylation and that AMPK activator blocked TPA-induced MMP-7 expression. In addition, a low concentration (5 μM) of H
2O
2 suppressed AMPK phosphorylation, which was prevented by DPI or antioxidants as well as siRNAs specific to NOX2 and p67phox. This result indicates that AMPK phosphorylation was suppressed by NOX2-activated ROS production in TPA-treated HT29 cells. Gene expression in response to NOX2-activated ROS production in HT29 cells was very similar to that in 5 μM H
2O
2-treated cells. Further, we observed for the first time that AMPK phosphorylation was differentially regulated according to ROS concentration, showing a U-shape dose–response curve associated with ATP production in HT29 cells. Low concentration of H
2O
2 up to 10 μM resulted in increased ATP production and a corresponding decrease in AMPK phosphorylation, which was reversed at above 10 μM H
2O
2. Our current study, which used a full range of H
2O
2 concentrations, actually supports previous findings showing that ROS activates AMPK [
34,
66,
67]. However, in contrast to previous reports of a direct relationship between AMPK activation and mitochondrial ROS production in cancer cells [
32], our results suggest that plasma membrane NOX2-derived ROS production deactivates AMPK. Despite our results that AMPK phosphorylation and ATP production are exactly matched in response to H
2O
2 treatment, the underlying mechanism must be elucidated.
AMPK phosphorylation was inversely correlated with the expression level of NOX2, indicating that there is a feedback loop between NOX and AMPK. However, AMPK activators alone in the absence of TPA did not suppress basal ROS production or NOX2 gene expression, suggesting that activation of NOX2 accompanied by ROS generation is a prerequisite of AMPK suppression. Further, our results suggest that deactivated AMPK did not exert any inhibitory effect on stimulus-induced NOX2 expression and thus MMP-7 gene induction over time.
Methods
Materials
All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA), if not specified. RPMI1640, MEM, fetal bovine serum (FBS), penicillin/streptomycin, and trizol reagent were purchased from Invitrogen Life Technologies (Carlsbad, CA, USA). Antibody of Lamin B, I-κB-α rabbit polyclonal antibody, β-actin mouse monoclonal antibody, NF-κB p65 rabbit polyclonal antibody, phospho-ERK antibody, MMP-7 goat polyclonal antibody, p67phox goat polyclonal antibody, NOX1 rabbit polyclonal antibody, NOX2 goat polyclonal antibody, celecoxib were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Phospho-p38 MAPK antibody, ERK antibody, p38 MAPK antibody, phospho-AMPK-α, AMPK-α, phospho-JNK, JNK, C-Jun, phospho I-κB, and PD98059 were purchased from Cell Signaling Technology, Inc. (Boston, MA, USA). Matrigel was obtained from BD Biosciences (Bedford, MA, USA). Trypsin/EDTA was purchased from Clonetics, Inc. (Walkersville, MD, USA). SR11302 was purchased from Tocris Bioscience (Tocris House, Bristol, BS110QL, UK). D942 was purchased from Calbiochem (10394, Pacific Center Ct, CA, USA).
Cell culture
Human colorectal cancer cells HT29 and SW620 were obtained from the American Type Culture Collection (Manassas, VA, USA). HT29 cells and SW620 cells were cultured in RPMI 1640 and DMEM high glucose media, respectively, containing 10 % FBS, 100 IU/ml of penicillin, and 100 μg/ml of streptomycin. Cells were maintained at 37 °C in 5 % CO2. After reaching 70 % confluence, cells were sub-cultured by splitting at 1:3 ratios.
Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was isolated using Trizol reagent following the manufacturer’s instructions. The extracted RNA was reverse-transcribed to cDNA using a GoScript Reverse Transcription system (#A5001, Promega Corporation, WI, USA). cDNA was amplified in the presence of 0.5 U of Taq DNA polymerase (Takara, Japan) using specific primers on Corbett Rotor-Gene (Corbett Life Science). Quantitative analysis of mRNA was done using a QuantiTect SYBR Green PCR kit (Qiagen). Primers for MMP-2, MMP-9, TIMP-1, TIMP-2, and GAPDH were supplied by Qiagen, and the other primer sequences used were MMP-7 (sense 5’-GGAGATGCTCACTTCGATGA-3’ and antisense 5’-ATACCCAAAGAATGGCCAAG -3’), NOX1 (sense 5′-GTTTTACCGCTCCCAGCAGA −3′ and antisense 5′-GGATGCCATTCCAGGAGAGA-3′), NOX2 (sense 5′-CCTAAGATAGCGGTTGATGG-3′ and antisense 5′-GACTTGAGAATGGATGCGAA -3′), p47phox (sense 5′-GCTGGTGGGTCATCAGGAAA-3′ and antisense 5′-GCCCTGACTTTTGCAGGTAC −3′), and p67phox (sense 5′-CCTGCAACTACCTTGAACCA −3′ and antisense 5′- GGACTGCGGAGAGCTTTCC-3′).
Matrigel invasion assay
Matrigel invasion assay was performed as described previously by Park
et al. [
46]. Briefly, inner and outer parts of the Transwell insert (BD Falcon, Franklin Lakes, USA) were coated with Matrigel (0.5 mg/ml) and collagen (1 mg/ml) respectively. TPA was added to all wells except the control well, after which 100 μl of cell suspension (5x10
5 cells/ml) in the presence or absence of inhibitors and antioxidants in serum-free media was added to the inner part of the transwell. After 24 h, invaded cells were fixed and stained with methanol and H&E separately. The number of invaded cells per field was captured using a microscope fitted with a camera at 200x magnification.
Subcellular fractionation and Western blot analysis
For subcellular fractionation, HT29 cells cultured in a 100-mm dish were serum-starved overnight and stimulated with TPA (0 to 60 min). Cells were harvested with 100 μl of subcellular fractionation buffer (SF) consisting of 250 mM sucrose, 20 mM HEPES (pH 7.4), 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 1 mM Dithiothreitol, 1x protease, and phosphatase inhibitor cocktail. Cells were passed through a 30 G needle 30–35 times on ice. The cell suspension was centrifuged at 700 g for 10 min at 4 °C. The supernatant was collected in a glass tube and centrifuged at 100,000 rpm (Sorvall RC M120 EX ultracentrifuge) for 1 h at 4 °C. After collecting supernatant containing the cytosolic fraction, the pellet was suspended in 60 μl of SF. The suspension was sonicated for 30 s four times at the highest setting with a 30-s break in each cycle. The suspension was centrifuged again at 100,000 rpm for 1 h at 4 °C, and supernatant containing the membrane fraction was collected.
Cytoplasmic and nuclear proteins were extracted using a NE-PER nuclear and cytoplasmic extraction reagent kit (#78833, Thermo Scientific, Rockford, USA) according to the manufacturer’s instructions. For isolation of total proteins, cells were lysed in radioimmunoprecipitation assay buffer (RIPA) containing protease and phosphatase inhibitor cocktail (Thermo Scientific). Cells were centrifuged at 12,000 rpm for 10 min, and supernatant containing soluble proteins was collected.
Protein concentration was determined using a BCA protein assay kit (Thermo Scientific). Equal quantities of proteins were resolved by 10-14 % SDS-PAGE, transferred onto a nitrocellulose membrane, and blocked using 5 % skim milk in TBS-Tween 20 (TBST) for 1 h. In the case of phospho-protein, membranes were blocked in 5 % bovine serum albumin in TBST for 1 h. Membranes were incubated with specific primary antibodies overnight at 4 °C, followed by incubation with horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. The immunoreactive proteins were detected using an enhanced chemiluminescent reagent (Thermo Scientific) system on a luminescent image analyzer, LAS-4000 mini (Fujifilm, Tokyo, Japan). In each step after blocking, membranes were washed with TBST.
Transfection of colon cancer cells with siRNA
HT29 and SW620 cells were transiently transfected using siRNAs (NOX1, NOX2, p67phox, and MMP-7) in opti-MEM I medium containing Dharmacon reagent (# T-2004-01, Thermo Scientific) as previously described by Regmi
et al. [
68]. The transfected cells were then subjected to invasion assay, ROS measurement, and Western blot analysis.
Lucigenin Chemiluminescence assay
Superoxide anion production was measured by lucigenin chemiluminescence assay as described by Regmi
et al. [
68] with some modifications. Cells were seeded in 96-well white opaque plates (1×10
5 cells/well). After overnight incubation, cells were pretreated with drugs in serum-free media prior to treatment with or without TPA (12 ng/ml) for the indicated time. Chemiluminescence was then measured using lucigenin (400 μM) with a Fluostar Optima microplate reader.
Intracellular ROS measurement
Cellular ROS was measured using a cell-permeable fluoregenic probe, 2’,7’-dichlorofluorescein diacetate (DCF-DA). Overnight serum-starved HT-29 cells (1×105 cells/cm2) were treated with or without TPA for the specific time points. After washing with PBS, cells were incubated with 10 μM DCF-DA at 37 °C for 30 min. The cells were washed again with PBS, and images were captured using a digital camera (TE2000-U, Nikon, Japan) with a blue filter (B-2E/C, FITC) at 200x magnification.
mRNA copy number determination
The number of transcripts of NOX1 and NOX2 was determined as previously described methods [
68,
69]. Briefly, human NOX1 and Nox2 cDNAs (Invitrogen) were cloned into pcDNA5/FRT/TO vector (Invitrogen). The following equation was used to calculate the copy number:
$$ \mathrm{D}\mathrm{N}\mathrm{A}\ \left(\mathrm{copies}\right) = \left[6.02 \times 1023\ \left(\mathrm{copies}/\mathrm{mol}\mathrm{e}\right) \times \mathrm{D}\mathrm{N}\mathrm{A}\ \mathrm{concentration}\ \left(\mathrm{g}\right)\right]\ /\ \left[\mathrm{D}\mathrm{N}\mathrm{A}\ \mathrm{length}\ \left(\mathrm{bp}\right) \times 660\ \left(\mathrm{g}/\mathrm{mol}/\mathrm{bp}\right)\right]. $$
To generate standard curve of NOX1 and NOX2, each plasmid were serially diluted 10 folds ranging from 101 to 105 copy numbers. cDNA synthesized from isolated total RNA of cells in the absence or presence of TPA for 24 h were subjected to real-time PCR using QuantiTect SYBER Green PCR kit (Qiagen) with NOX1 and Nox2 primer sequences. Copy number of NOX1 and Nox2 in HT29, Caco-2, SW620, HCT116 was calculated from the standard curve.
NADPH Oxidase activity assay
The NADPH oxidase activity assay was performed by modifying the method described by [
70]. Briefly, cells were transfected with siRNAs of non-specific, NOX1 and NOX2. The cells were harvested in Krebs-HEPES buffer (pH 7.4) containing protease and phosphatase inhibitor cocktail (Thermo Scientific), homogenized with Dounce homogenizer, and centrifuged at 10,000 g for 15 min. After determining protein concentration using a BCA protein assay kit (Thermo Scientific), equal amount of proteins was transferred to the 96 well (white plate) with 10 μM lucigenin prepared in the same buffer and incubated at 37 °C for 10 min. Then, 100 μM NADPH was added to each wells with or without TPA. After 5 min incubation at 37 °C, the chemiluminescence was measured with a Fluostar Optima microplate reader.
Measurement of ATP
Mitochondrial function was assessed using Mitochondrial ToxGlo™ Assay (#G8000, Promega). Briefly, HT-29 cells (10,000 cells/well) were seeded in a 96-well opaque white-walled flat bottom plate (Falcon). After 24 h of incubation at 37 °C, cells were washed with PBS, and serum-free galactose-containing media (glucose-free) was added to each well. Cells were treated with different concentrations of H2O2 (1 to 500 μM) or TPA (12 ng/ml) for 1 h. Cytotoxicity was first assessed using fluorogenic peptide substrate (bis-AAF-R110). Fluorescence was measured using a Fluostar optima microplate reader (BMG Labtech GmbH, Offenburg, Germany) with excitation at 485 nm and emission at 520 nm. Next, cells were lysed by addition of ATP detection reagent, and luminescence was measured using a Fluostar optima microplate reader following the manufacturer’s instructions from the Mitochondrial ToxGlo™ assay kit. The amount of ATP is directly proportional to the luminescence signal.
Reporter plasmid transfection and Luciferase activity measurement
Transactivation of NF-κB and AP-1 was studied using the dual Luciferase reporter assay system (Promega, Madison, WI) following the manufacturer’s instructions. Briefly, HT29 cells were seeded in a 24-well plate at a density of 7×104 cells/well in antibiotic-free media containing 10 % FBS. The next day, cells were co-transfected with NF-κB (Affymetrix) or AP-1 (Panomics) reporter vectors along with control vector (pRL-TK) using Lipofectamine 2000. The transfection media was replaced with RPMI 1640 containing 10 % FBS after 18 h, and the cells were further allowed to grow for another 24 h. Then, the transfected cells were treated with MAPK inhibitors for 1 h prior to treatment with TPA for another 3 h. The lysates were used for the assay.
Statistical analysis
All data are the mean of three independent experiments. Error bar represents SEM Statistical significance was determined using Student’s t-test or one-way ANOVA followed by the Student-Newman-Keuls comparison method for calculation of differences between groups (GraphPad Prism 5.0 software, San Diego, CA, USA). Values of P < 0.05 were considered statistically significant.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
There is no conflict of interest.
Authors’ contributions
Study conception and design: J-AK. Acquisition of data: SB, SCR. Analysis and interpretation of data: SB, SCR, J-AK. Drafting of manuscript: SB, J-AK. Critical revision: J-AK. All authors read and approved the final manuscript.